Human assisted positioning using textual signs

Bo Han, Feng Qian, Moo Ryong Ra

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

Location information is one of the key enablers to context-aware systems and applications for mobile devices. However, most existing location sensing techniques do not work or will be significantly slowed down without infrastructure support, which limits their applicability in several cases. In this paper, we propose a localization system that works for both indoor and outdoor environments in a completely offline manner. Our system leverages human users'perception of nearby textual signs, without using GPS, Wi-Fi, cellular, and Internet. It enables several important use cases, such as offline localization on wearable devices. Based on real data collected from Google Street View and OpenStreetMap, we examine the feasibility of our approach. The preliminary result was encouraging. Our system was able to achieve higher than 90% accuracy with only 4 iterations even when the speech recognition accuracy is 70%, requiring very small storage space, and consuming 44% less instantaneous power compared to GPS.

Original languageEnglish (US)
Title of host publicationHotMobile 2015 - 16th International Workshop on Mobile Computing Systems and Applications
PublisherAssociation for Computing Machinery, Inc
Pages87-92
Number of pages6
ISBN (Electronic)9781450333917
DOIs
StatePublished - Feb 12 2015
Externally publishedYes
Event16th International Workshop on Mobile Computing Systems, HotMobile 2015 - Santa Fe, United States
Duration: Feb 12 2015Feb 13 2015

Publication series

NameHotMobile 2015 - 16th International Workshop on Mobile Computing Systems and Applications

Conference

Conference16th International Workshop on Mobile Computing Systems, HotMobile 2015
CountryUnited States
CitySanta Fe
Period2/12/152/13/15

Fingerprint Dive into the research topics of 'Human assisted positioning using textual signs'. Together they form a unique fingerprint.

Cite this